Dry sprinkler systems are well-known in the art. A dry sprinkler system includes a sprinkler grid having a plurality of sprinkler heads. The sprinkler grid is connected via fluid flow lines containing air or other gas. The fluid flow lines are coupled to a primary water supply valve which can include, for example, an air-to-water ratio valve, deluge valve or preaction valve as is known in the art. The sprinkler heads typically include normally closed temperature-responsive valves. The normally closed valves of the sprinkler heads open when sufficiently heated or triggered by a thermal source such as a fire. The open sprinkler head, alone or in combination with a smoke or fire indicator, causes the primary water supply valve to open, thereby allowing the service water to flow into the fluid flow lines of the dry pipe sprinkler grid (displacing the air therein), and through the open sprinkler head to control the fire, reduce the smoke source, and/or minimize any damage therefrom. Water flows through the system and out the open sprinkler head (and any other sprinkler heads that subsequently open), until the sprinkler head closes itself, if automatically resetting, or until the water supply is turned off.
In contrast, a wet pipe sprinkler system has fluid flow lines that are pre-filled with water. The water is retained in the sprinkler grid by the valves in the sprinkler heads. As soon as a sprinkler head opens, the water in the sprinkler grid immediately flows out of the sprinkler head. In addition, the primary water valve in the wet sprinkler system is the main shut-off valve, which is in the normally open state.
There are three types of dry sprinkler systems that contain air or gas as opposed to water or other fluid. These dry systems include: dry pipe, preaction, and deluge systems. A dry pipe system includes fluid flow pipes which are charged with air under pressure and when the dry pipe system detects heat from a fire, the sprinkler heads open resulting in a decrease in air pressure. The resultant decrease in air pressure activates the water supply source and allows water to enter the piping system and exit through the sprinkler heads.
In a deluge system, the fluid flow pipes remain free of water, employs sprinkler heads that remain open, and utilizes pneumatic or electrical detectors to detect an indication of fire such as, for example, smoke or heat. The network of pipes in a deluge system usually do not contain supervisory air, but will instead contain air at atmospheric pressure. Once the pneumatic or electrical detectors detect heat, the water supply source provides water to the pipes and sprinkler heads. A preaction system has pipes that are free of water, employs sprinkler heads that remain closed, has supervisory air, and utilizes pneumatic or electrical detectors to detect an indication of fire such as, for example, heat or smoke. Only when the system detects a fire is water introduced into the otherwise dry network of pipes and sprinkler heads.
When a dry pipe sprinkler system goes “wet” (i.e., to cause the primary water supply valve to open and allow the water to fill the fluid flow supply lines), a sprinkler head opens, the pressure difference between the air pressure in the fluid flow lines and the water supply pressure on the wet side of the primary water supply valve or dry pipe air-to-water ratio valve reaches a specific hydraulic/pneumatic imbalance to open up the valve and release the water supply into the network of pipes. It may take up to 120 seconds to reach this state, depending upon the volume of the entire sprinkler system, water supply and air pressure. The larger the water supply, the larger the air supply is needed to hold the air-to-water ratio valve closed. Moreover, if the system is large and/or if the system is charged to a typical pressure such as 40 psig, a considerable volume of air must escape or be expelled from the open sprinkler head before the specific hydraulic imbalance is reached to open the primary water valve. The water supply travels through the piping grid displacing the pressurized gas to finally discharge through the open sprinkler.
The travel time of both the escaping gas and the fluid supply through the network provides for a fluid delivery delay in dry sprinkler systems that is not present in wet sprinkler systems. Currently, there exists an industry-wide belief that in dry sprinkler systems it is best to minimize or if possible, avoid fluid delivery delay. This belief has led to an industry-wide perception that dry sprinkler systems are inferior to wet systems. Current industry accepted design standards attempt to address or minimize the impact of the fluid delivery delay by placing a limit on the amount of delay that can be in the system. For example, NFPA-13, at Sections 7 and 11 that the water must be delivered from the primary water control valve to discharge out of the sprinkler head at operating pressure in under sixty seconds and more specifically under forty seconds. To promote the rapid delivery of water in dry sprinkler systems, Section 7 of the NFPA-13 further provides that, for dry sprinkler systems having system volumes between 500 and 750 gallons, the discharge time-limit can be avoided provided the system includes quick-opening devices such as accelerators.
The NFPA standards provide other various design criteria for both wet and dry sprinkler systems used in storage occupancies. Included in NFPA-13 are density-area curves and density-area points that define the requisite discharge flow rate of the system over a given design area. A density-area curve or point can be specified or limited in system design for protection of a given type of commodity classified by class or by groups as set forth in NFPA-13—Sections 5.6.3 and 5.6.4. For example, NFPA-13 provides criteria for the following commodity classes: Class I; Class II; Class III and Class IV. In addition, NFPA-13 provides criteria for the following groups to define the groups of plastics, elastomers or rubbers as Group A; Group B; and Group C.
NFPA-13 provides for additional provisions in the design of dry protection systems used for protecting stored commodities. For example, NFPA requires that the design area for a dry sprinkler system be increase in size as compared to a wet systems for protection of the same area or space. Specifically, NFPA-13—Section 12.1.6.1 provides that the area of sprinkler operation, the design area, for a dry system shall be increased by 30 percent (without revising the density) as compared to an equivalent wet system. This increase in sprinkler operational area establishes a “penalty” for designing a dry system; again reflecting an industry belief that dry sprinkler systems are inferior to wet.
For protection of some storage commodities, NFPA-13 provides design criteria for ceiling-only sprinkler systems in which the design “penalty” is greater than thirty percent. For example, certain forms of rack storage require a dry ceiling sprinkler system to be supplemented or supported by in-rack sprinklers as are known in the art. A problem with the in-rack sprinklers are that they may be difficult to maintain and are subject to damage from forklifts or the movement of storage pallets. NFPA-13 does provide in NFPA-13—Section 12.3.3.1.5; FIG. 12.3.3.1.5(e), Note 4, standards for protection of Group A plastics using a dry ceiling-only system having appropriately listed K-16.8 sprinklers for ceilings not exceeding 30 ft. in height. The design criteria for ceiling only storage wet sprinkler system is 0.8 gpm/ft2 per 2000 ft2. However, NFPA adds an additional penalty for dry system ceiling-only sprinkler systems by increasing the design criteria to 0.8 gpm/ft2 per 4500 ft2. This increased area requirement is a 125% density penalty over the wet system design criteria. As noted, the design penalties of NFPA-13 are believed to be provided to compensate for the inherent fluid delivery delay in a dry sprinkler system following thermal sprinkler activation. Moreover, NFPA 13 provides limited ceiling-only protection in limited rack storage configurations, and otherwise require in-rack sprinklers.
In complying with the thirty percent design area increase and other “penalties”, fire protection system engineers and designers are forced to anticipate the activation of more sprinklers and thus perhaps provide for larger piping to carry more water, larger pumps to properly pressurize the system, and larger tanks to make-up for water demand not satisfied by the municipal water supply. Despite the apparent economic design advantage of wet systems over dry systems, certain storage configurations prohibit the use of wet systems or make them otherwise impractical. Dry sprinkler systems are typically employed for the purpose of providing automatic sprinkler protection in unheated occupancies and structures that may be exposed to freezing temperatures. For example, in warehouses using high rack storage, i.e. 25 ft. high storage beneath a 30 ft. high ceiling, such warehouses may be unheated and therefore susceptible to freezing conditions making wet sprinkler systems undesirable. Freezer storage presents another environment that cannot utilize wet systems because water in the piping of the fire protection system located in the freezer system would freeze. One solution to the problem that has been developed is to use sprinklers in combination with antifreeze. However, the use of antifreeze can raise other issues such as, for example, corrosion and leakage in the piping system. In addition, the high viscosity of antifreeze may require increased piping size. Moreover, propylene glycol (PG) antifreeze has been shown not to have the fire-fighting characteristics of water and in some instances has been known to momentarily accelerate fire growth.
Generally, dry sprinkler systems for storage occupancies are configured for fire control in which a fire is limited in size by the distribution of water from one or more thermally actuated sprinkler located above the fire to decrease the heat release rate and pre-wet adjacent combustibles while controlling ceiling gas temperatures to avoid structural damage. However, with this mode of addressing a fire, hot gases may be entrained or maintained in the ceiling area above the fire and allowed to migrate radially. This may result in additional sprinklers being activated remotely from the fire and thus not impact the fire directly. In addition, the discharge of fluid from a given sprinkler can result in the impingement of water droplets and/or the build up of condensation of water vapor on adjacent and unactuated sprinklers. The resultant effect of unactuated sprinklers inter-dispersed between actuated sprinklers is known as sprinkler skipping. One definition of sprinkler skipping is the “significantly irregular sprinkler operating sequence when compared to the expected sequence dictated by the ceiling flow behavior, assuming no sprinkler system malfunctions.” See PAUL. A. CROCE ET AL., An Investigation of the Causative Mechanism of Sprinkler Skipping, 15 J. FIRE PROT. ENGR. 107, 107 (May 2005). Due to the actuation of additional remote sprinklers, current design criteria may require enlarged piping, and thus, the volume of water discharge into the storage area may be larger than is adequately necessary to address the fire. Moreover, because fire control merely reduces heat release rate, a large number of sprinkles may be activated in response to the fire in order to maintain the heat release rate reduction.
Despite the availability of immediate fluid delivery from each sprinkler in a wet sprinkler system, wet sprinkler systems can also experience sprinkler skipping. However, wet sprinkler systems can be configured for fire suppression which sharply reduces the heat release rate of a fire and prevents its regrowth by means of direct and sufficient application of water through the fire plume to the burning fuel surface. For example, a wet system can be configured to use early suppression fast-response (ESFR) Sprinklers. The use of ESFR sprinklers is generally not available in dry sprinklers systems, to do so would require a specific listing for the sprinkler as is required under Section 8.4.6.1 of NFPA-13. Thus, to configure a dry sprinkler system for fire suppression may require overcoming the additional penalty of a specific listing for an ESFR sprinkler. Moreover, to hydraulically configure a dry system for suppression may require adequately sized piping and pumps whose costs may prove economically prohibitive as these design constraints may require hydraulically sizing the system beyond the demands already imposed by the design “penalties.”
Two fire tests were conducted to determine the ability of a tree-type dry pipe or double-interlock preaction system employing ceiling-only Large Drop sprinklers to provide adequate fire protection for rack storage of Class II commodity at a storage height of thirty-four feet (34 ft.) beneath a ceiling having a ceiling height of forty feet. One fire test showed that the system, employing a thirty second (30 sec.) or less water delay time, could provide adequate fire control with a discharge water pressure of 55 psi. However, in addition to the high operating pressure of 55 psi., such a system required a total of twenty-five (25) sprinkler operations actuated over a seventeen minute period. The second fire test employed a sixty-second (60 sec.) water delay time, however such a delay time proved to be too long as the fire developed to such a severity that adequate fire control could not be achieved. In the second fire test, seventy-one (71) sprinklers operated resulting in a maximum discharge pressure of 37 psi., and thus, the target pressure of 75 psi. could not be attained. The tests and their results are described in Factory Mutual Research Technical Report: FMRC J.I. 0Z0R6.RR NS entitled, “Dry Pipe Sprinkler Protection of Rack Stored Class II Commodity In 40-Ft. High Buildings,” prepared for Americold Corp. and published June 1995.
In an attempt to understand and predict fire behavior, The National Institute of Standards and Technology (NIST) has developed a software program entitled Fire Dynamics Simulator (FDS), currently available from the NIST website, Internet<URL: http://fire.nist.gov/fds/, that models the solution of fire driven flows, i.e. fire growth, including but not limited to flow velocity, temperature, smoke density and heat release rate. These variables are further used in the FDS to model sprinkler system response to a fire.
FDS can be used to model sprinkler activation or operation of a dry sprinkler system in the presence of a growing fire for a stored commodity. One particular study has been conducted using FDS to predict fire growth size and the sprinkler activation patterns for two standard commodities and a range of storage heights, ceiling heights and sprinkler installation locations. The findings and conclusions of the study are discussed in a report by David LeBlanc of Tyco Fire Products R&D entitled, Dry Pipe Sprinkler Systems—Effect of Geometric Parameters on Expected Number of Sprinkler Operation (2002) hereinafter “FDS Study”) which is incorporated in its entirety by reference.
The FDS Study evaluated predictive models for dry sprinkler systems protecting storage arrays of Group A and Class II commodities. The FDS Study generated a model that simulated fire growth and sprinkler activation response. The study further verified the validity of the prediction by comparing the simulated results with actual experimental tests. As described in the FDS study, the FDS simulations can generate predictive heat release profiles for a given stored commodity, storage configuration and commodity height showing in particular the change in heat release over time and other parameters such as temperature and velocity within the computational domain for an area such as, for example, an area near the ceiling. In addition, the FDS simulations can provide sprinkler activation profiles for the simulated sprinkler network modeled above the commodity showing in particular the predicted location and time of sprinkler activation.